Biometry
Biology 325
Instructor:
Jerry G. Chmielewski
Objectives: An introduction to statistical techniques and experimental
design as applied to environmental or biological problems. Emphasis
is placed on the selection and interpretation of tests, rather than
theory. Descriptive methods, tests of significance, linear
regression, correlation, analysis of variance and covariance, and
non-parametric techniques are included. The laboratory component of
the course provides a thorough introduction to the use of computers
in statistics. Considerable attention is devoted to the use of PC
SAS for data analysis. One section of this course is offered in the
spring semester.
Prerequisites: None
Credit
Value: 3
Contact
Time: Two 50
minute lectures and one 3 hour laboratory per week.
Outcomes:
Upon completion of this
course a student should:
- understand the meaning
of and be able to calculate the mean, confidence intervals for the
mean, standard deviation, standard error, product moment
correlation coefficients, slope of the least-squares regression
line, Chi-square, etc.
- understand hypothesis
testing and be able to construct a null and an alternate
hypothesis
- understand the meaning
of statistical significance as it relates to a probability
statement, and be able to correctly reject or fail to reject test
hypotheses
- understand the normal
distribution, the student's distribution, the binomial
distribution, the Poisson distribution, the Chi-square
distribution, and the F distribution
- know when and how to
use each of the above distributions for testing
hypotheses
- be able to test the
difference between correlated and uncorrelated sample means using a
t-test for two means and analysis of variance for several
means
- understand the meaning
of and be able to test homogeneity of variances
- understand how to probe
for significant differences between individual means after a
significant ANOVA
- understand simple ANOVA
designs (i.e. randomized, blocked, nested, Latin square,
factorial)
- be able to express the
relationship between two variables by correlation
analysis
- be able to analyze the
dependence of one variable upon another by regression
- understand the concepts
of multiple correlation and regression
- understand the concept
of covariate analysis
- be able to test for
goodness of fit of data to various genetic or other a priori models
by Chi-square analysis
- be able to construct
contingency tables to test for association between
variables
- understand the
necessity for and use of non-parametric tests and should be
familiar with one or more appropriate non-parametric tests for
differences between samples, correlation, and analysis of
variance
- have a basic
understanding of the necessity to include statistical planning in
research design
- be able to read
articles in the primary literature and critique the respective
authors experimental design, use of ststatistical procedures, and
interpretation of results
- be versed in the use of
PC SAS
Assessment:
- Term test 1: Questions
will be comprehensive dealing with material covered during
scheduled lecture and laboratory periods. The test will consist of
two parts. Part 1 will be in class and deal with the technical
aspects of statistical testing. Part 2 will be out of class, open
book, and restricted to problem solving. You will need to be
familiar with PC SAS to answer these question. (15
Percent)
- Term test 2: Questions
will be comprehensive dealing with material covered during
scheduled lecture and laboratory periods. The test will consist of
two parts. Part 1 will be in class and deal with the technical
aspects of statistical testing. Part 2 will be out of class, open
book, and restricted to problem solving. You will need to be
familiar with PC SAS to answer these question. (20
Percent)
- Assignments: Questions
will likely be assigned on a weekly basis during the laboratory
period. These questions will be comprehensive. Critiques of
published works will be considered as part of this grade. (30
Percent)
- Final examination:
Questions will be comprehensive dealing with material covered
during scheduled lecture and laboratory periods. The test will
consist of two parts. Part 1 will be in class and deal with the
technical aspects of statistical testing. Part 2 will be out of
class, open book, and restricted to problem solving. You will need
to be familiar with PC SAS to answer these question. (35
Percent)
Lecture
Schedule:
- Syllabus,
biostatistics, data
- Frequency
distributions, populations, samples
- Measures of central
tendency
- Measures of
dispersion
- Hypotheses, alpha,
chi-square
- Yates correction for
continuity, heterogeneity chi-square
- Log-likelihood,
Kolmogorov-Smirnov
- Contingency tables
(2X2), subdividing contingency tables
- Normal
distribution
- Coding, data
transformations
- t-test, one and two
tailed tests, confidence limits
- Two sample hypotheses,
parametric tests
- Mann Whitney non
parametric test
- Paired sample t-test,
Wilcoxon paired sample test
- Single factor analysis
of variance
- Single factor analysis
of variance
- Kruskal-Wallis non
parametric test
- Nested analysis of
variance
- Multiple range
testing
- Two factor analysis of
variation with replication
- Two factor analysis of
variation with replication
- Two factor analysis of
variation without replication
- Multifactorial analysis
of variance, Latin square, blocked designs
- Linear
regression
- Comparing simple linear
regressions
- Correlation
analysis
- Analysis of
covariance
- Analysis of
covariance
- Binomial
distribution
- Poisson
distribution
Laboratory
Schedule:
- The laboratory
component of the course provides a thorough introduction to the use
PC SAS for the purpose of data analysis.
Miscellaneous Related
Links:
|